One class of opto-electrical devices is that using an organic material for light emission (or detection in the case of photovoltaic cells and the like). The basic structure of these devices is a light emissive organic layer, for instance a film of a poly (p-phenylenevinylene) (“PPV”) or polyfluorene, sandwiched between a cathode for injecting negative charge carriers (electrons) and an anode for injecting positive charge carriers (holes) into the organic layer. The electrons and holes combine in the organic layer generating photons. In WO90/13148 the organic light-emissive material is a polymer. In U.S. Pat. No. 4,539,507 the organic light-emissive material is of the class known as small molecule materials, such as (8-hydroxyquinoline) aluminum (“Alq3”). In a practical device one of the electrodes is transparent, to allow the photons to escape the device.
A typical organic light-emissive device (“OLED”) is fabricated on a glass or plastic substrate coated with a transparent anode such as indium-tin-oxide (“ITO”). A layer of a thin film of at least one electroluminescent organic material covers the first electrode. Finally, a cathode covers the layer of electroluminescent organic material. The cathode is typically a metal or alloy and may comprise a single layer, such as aluminum, or a plurality of layers such as calcium and aluminum.
In operation, holes are injected into the device through the anode and electrons are injected into the device through the cathode. The holes and electrons combine in the organic electroluminescent layer to form an exciton which then undergoes radiative decay to give light (in light detecting devices this process essentially runs in reverse).
OLEDs can provide a particularly advantageous form of electro-optic display. They are bright, colorful, fast-switching, provide a wide viewing angle and are easy and cheap to fabricate on a variety of substrates. Organic (which here includes organometallic) LEDs may be fabricated using polymers, dendrimers, and/or small molecules in a range of colors, depending upon the materials used.
OLEDs may be deposited on a substrate in a matrix of pixels to form a single or multi-color pixelated display. A multi-colored display may be constructed using groups of red, green, and blue emitting pixels. So-called active matrix displays have a memory element, typically a storage capacitor and a transistor, associated with each pixel whilst passive matrix displays have no such memory element and instead are repetitively scanned to give the impression of a steady image.
FIG. 1 shows a vertical cross section through an example of an OLED device 100. In an active matrix display, part of the area of a pixel is occupied by associated drive circuitry (not shown in FIG. 1). The structure of the device is somewhat simplified for the purposes of illustration.
The OLED 100 comprises a substrate 102, typically 0.7 mm or 1.1 mm glass but optionally clear plastic, on which an anode layer 106 has been deposited. The anode layer typically comprises around 150 nm thickness of ITO (indium tin oxide), over which is provided a metal contact layer, typically around 500 nm of aluminum, sometimes referred to as anode metal. Glass substrates coated with ITO and contact metal may be purchased from Corning, USA. The contact metal (and optionally the ITO) is patterned as desired so that it does not obscure the display, by a conventional process of photolithography followed by etching.
A substantially transparent hole transport layer 108a is provided over the anode metal, followed by an electroluminescent layer 108b. Banks 112 may be formed on the substrate, for example from positive or negative photoresist material, to define wells 114 into which these active organic layers may be selectively deposited, for example by a droplet deposition or inkjet printing technique. The wells thus define light emitting areas or pixels of the display.
A cathode layer 110 is then applied by, say, physical vapor deposition. The cathode layer typically comprises a low work function metal such as calcium or barium covered with a thicker, capping layer of aluminum and optionally including an additional layer immediately adjacent the electroluminescent layer, such as a layer of lithium fluoride, for improved electron energy level matching. The cathode may be transparent. This is particularly preferred for active matrix devices wherein emission through the substrate is partially blocked by drive circuitry located underneath the emissive pixels. In the case of a transparent cathode device, it will be appreciated that the anode is not necessarily transparent. In the case of passive matrix displays, mutual electrical isolation of cathode lines may achieved through the use of cathode separators (element 302 of FIG. 3b). Typically a number of displays are fabricated on a single substrate and at the end of the fabrication process the substrate is scribed, and the displays separated. An encapsulant such as a glass sheet or a metal can is utilized to inhibit oxidation and moisture ingress.
OLEDs of this general type may be fabricated using a range of materials including polymers, dendrimers, and so-called small molecules, to emit over a range of wavelengths at varying drive voltages and efficiencies. Examples of polymer-based OLED materials are described in WO 90/13148, WO 95/06400 and WO 99/48160; examples of dendrimer-based materials are described in WO 99/21935 and WO 02/067343; and examples of small molecule OLED materials are described in U.S. Pat. No. 4,539,507. The aforementioned polymers, dendrimers and small molecules emit light by radiative decay of singlet excitons (fluorescence). However, up to 75% of excitons are triplet excitons which normally undergo non-radiative decay. Electroluminescence by radiative decay of triplet excitons (phosphorescence) is disclosed in, for example, “Very high-efficiency green organic light-emitting devices based on electrophosphorescence” M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, and S. R. Forrest Applied Physics Letters, Vol. 75(1) pp. 4-6, Jul. 5, 1999″. In the case of a polymer-based OLED, layers 108 comprise a hole injection layer 108a and a light emitting polymer (LEP) electroluminescent layer 108b. The electroluminescent layer may comprise, for example, around 70 nm (dry) thickness of PPV (poly(p-phenylenevinylene)) and the hole injection layer, which helps match the hole energy levels of the anode layer and of the electroluminescent layer, may comprise, for example, around 50-200 nm, preferably around 150 nm (dry) thickness of PEDOT:PSS (polystyrene-sulphonate-doped polyethylene-dioxythiophene).
FIG. 2 shows a view from above (that is, not through the substrate) of a portion of a three-color active matrix pixelated OLED display 200 after deposition of one of the active color layers. The figure shows an array of banks 112 and wells 114 defining pixels of the display.
FIG. 3a shows a view from above of a substrate 300 for inkjet printing a passive matrix OLED display. FIG. 3b shows a cross-section through the substrate of FIG. 3a along line Y-Y′.
Referring to FIGS. 3a and 3b, the substrate is provided with a plurality of cathode undercut separators 302 to separate adjacent cathode lines (which will be deposited in regions 304). A plurality of wells 308 is defined by banks 310, constructed around the perimeter of each well 308 and leaving an anode layer 306 exposed at the base of the well. The edges or faces of the banks are tapered onto the surface of the substrate as shown, heretofore at an angle of between 10 and 40 degrees. The banks present a hydrophobic surface in order that they are not wetted by the solution of deposited organic material and thus assist in containing the deposited material within a well. This may be achieved by treatment of a bank material such as polyimide with an O2/CF4 plasma as disclosed in EP 0989778. Alternatively, the plasma treatment step may be avoided by use of a fluorinated material such as a fluorinated polyimide as disclosed in WO 03/083960.
As previously mentioned, the bank and separator structures may be formed from resist material, for example using a positive (or negative) resist for the banks and a negative (or positive) resist for the separators; both these resists may be based upon polyimide and spin coated onto the substrate, or a fluorinated or fluorinated-like photoresist may be employed. In the example shown the cathode separators are around 5 μm in height and approximately 20 μm wide. Banks are generally between 20 μm and 100 μm in width and in the example shown have a 4 μm taper at each edge (so that the banks are around 1 μm in height). The pixels of FIG. 3a are approximately 300 μm square but, as described later, the size of a pixel can vary considerably, depending upon the intended application.
The deposition of material for organic light emitting diodes (OLEDs) using ink jet printing techniques is described in a number of documents including, for example: T. R. Hebner, C. C. Wu, D. Marcy, M. H. Lu and J. C. Sturm, “Ink-jet Printing of doped Polymers for Organic Light Emitting Devices”, Applied Physics Letters, Vol. 72, No. 5, pp. 519-521, 1998; Y. Yang, “Review of Recent Progress on Polymer Electroluminescent Devices,” SPIE Photonics West: Optoelectronics '98, Conf. 3279, San Jose, January, 1998; EP 0 880 303; and “Ink-Jet Printing of Polymer Light-Emitting Devices”, Paul C. Duineveld, Margreet M. de Kok, Michael Buechel, Aad H. Sempel, Kees A. H. Mutsaers, Peter van de Weijer, Ivo G. J. Camps, Ton J. M. van den Biggelaar, Jan-Eric J. M. Rubingh and Eliav I. Haskal, Organic Light-Emitting Materials and Devices V, Zakya H. Kafafi, Editor, Proceedings of SPIE Vol. 4464 (2002). Ink jet techniques can be used to deposit materials for both small molecule and polymer LEDs.
A volatile solvent is generally employed to deposit a molecular electronic material, with 0.5% to 4% dissolved material. This can take anything between a few seconds and a few minutes to dry and results in a relatively thin film in comparison with the initial “ink” volume. Often multiple drops are deposited, preferably before drying begins, to provide sufficient thickness of dry material. Typical solvents which have been used include cyclohexylbenzene and alkylated benzenes, in particular toluene or xylene; others are described in WO 00/59267, WO 01/16251 and WO 02/18513; a solvent comprising a blend of these may also be employed. Precision ink jet printers such as machines from Litrex Corporation of California, USA are used; suitable print heads are available from Xaar of Cambridge, UK and Spectra, Inc. of NH, USA.
The last few years have seen an increasing activity in the development of ink jet printing for depositing electronic materials. In particular there have been demonstrations of ink jet printing of charge injection, charge transport and electroluminescent layers of OLED devices by more than a dozen display manufacturers.
The key reasons for the interest in ink jet printing are scalability and adaptability. The former allows arbitrarily large sized substrates to be patterned and the latter should mean that there are negligible tooling costs associated with changing from one product to another since the image of dots printed on a substrate is defined by software. At first sight this would be similar to printing a graphic image—commercial print equipment is available that allow printing of arbitrary images on billboard sized substrates [Inca digital website: http://www.incadigital.com/]. However the significant difference between graphics printers and display panels is that the former use substrates that are porous or use inks that are UV curable resulting in very little effect of the drying environment on film formation. In comparison, the inks used in fabricating OLED displays are ink jet printed onto non-porous surfaces and the process of changing from a wet ink to dry film is dominated by the drying environment of the ink in the pixel. In addition OLED devices require the films to be uniform to nanometer tolerance. It follows that to achieve scalability and adaptability requires control of the film forming properties of the ink and a robustness of this process to changes in pixel dimensions and swathe timing.
The aforementioned control is far from a simple matter as the physics of the interactions of a drying solution with a surface are extremely complicated and a complete theory still awaits development. In general terms, the behavior of drying drops of ink is explained by the coffee-ring effect first modeled by Deegan [R. D. Deegan, O. Bakajin, T. F. Dupont, G. Huber, S. R. Nagel, and T. A. Witten. Capillary flow as the cause of ring stains from dried liquid drops. Nature 389, 827 (1997)]. For the case of circular pixels the wet ink forms a section of a sphere, where the angle made by the drop surface with the substrate is the contact angle. When pinning occurs (which it invariably does for the inks and surfaces used in polymer OLED display manufacturing) the drying drop maintains its diameter and solute is carried to the edges of the drop forming a ring of material at the outer edges of the pixel. The amount of material carried to the edge depends on a number of factors—in particular how long the process of material transfer can occur before the drying drop gels and the uniformity of the drying environment.
In some circumstance, the dissolved molecular electronic material deposited into a well may dry to form a film with a relatively thin edge. FIGS. 4a and 4b illustrate this process. FIG. 4a shows a simplified cross section 400 through a well 308 filled with dissolved material 402, and FIG. 4b shows the same well after the material has dried to form a solid film 404. In this example the bank angle is approximately 15° and the bank height is approximately 1.5 μm. As can be seen a well is generally filled until it is brimming over. The solution 402 has a contact angle θc with the plasma treated bank material of typically between 30° and 40° for example around 35°; this is the angle the surface of the dissolved material 402 makes with the (bank) material it contacts, for example angle 402a in FIG. 4a. As the solvent evaporates the solution becomes more concentrated and the surface of the solution moves down the tapering face of a bank towards the substrate; pinning of the drying edge can occur at a point between the initially landed wet edge and the foot of the bank (base of the well) on the substrate. The result, shown in FIG. 4b, is that the film of dry material 404 can be very thin, for example of the order of 10 nm or less, in a region 404a where it meets the face of a bank.
In other circumstance, due to the coffee ring effect the dissolved molecular electronic material deposited into a well may dry to form a film with a relatively thick edge. Because the thickness of solution is less at the edge of a drop than in the center, as the edge dries the concentration of dissolved material there increases. Because the edge tends to be pinned solution then flows from the centre of the drop towards the edge to reduce the concentration gradient. This effect can result in dissolved material tending to be deposited in a ring rather than uniformly.
In addition to the above problems, since the printing process involves printing stripes (or swathes) of ink (corresponding to the ink jet head width) there is an inbuilt asymmetry in the drying environment. At a swathe edge more drying occurs on the unprinted side since the solvent concentration in the atmosphere above the substrate is less than the printed side. With more evaporation taking place on the unprinted side more solute is deposited on this side and the film profile becomes asymmetric.
Yet another problem is that an inkjet droplet may not fall exactly into a well but instead land in part on the slope of the bank and may dry in place, resulting in non-uniformities in the end display.
A further problem with inkjet deposition arises when filling wells which are large compared with the size of an inkjet droplet. A typical droplet from an inkjet print head has a diameter of approximately of 30 μm in flight and the droplet grows to approximately 100 μm in diameter when it lands and wets out. However it is difficult to produce drops of, say 100 μm in diameter (in flight) from a print head. As such, several droplets may be required to fill a large well. FIG. 5a which shows a well 500 for a long thin pixel of a type which is typically used in a RGB (red green blue) display. In the example of FIG. 5a the pixel has a width of 50 μm and a length of 150 μm with 20 μm wide banks (giving a 70 μm pixel pitch and a 210 μm full color pitch). Such a well can be filled by three 50 μm droplets 502a, b, c as shown. Referring now to FIG. 5b this shows a well 510 for a pixel which is approximately four times larger than each dimension giving a pixel width of approximately 200 μm, more suitable for applications such as a color television. As can be seen from the figure, many droplets 512 are needed to fill such a pixel. In practice, these tend to coalesce to form a larger droplet 514 which tends not to properly fill corners of the pixel (although FIGS. 5a and 5b are idealized and, in practice, the corners are not generally as sharp as they are shown). One way around this problem is to over fill the well to a sufficient extent that the dissolved material well is pushed into the corners. This can be achieved by using a large number of dilute droplets and a high barrier around the well. Techniques for depositing large volumes of liquid are described in WO 03/065474, which describes the use of very high barriers (for examples at page 8 lines 8 to 20) to allow the wells to hold a large volume of liquid without the liquid overflowing to adjacent wells. However such structures cannot easily be formed by photolithography and instead a plastic substrate is embossed or injection molded. It is also desirable to be able to fill a well using fewer (higher concentration) droplets as this enables, inter alia faster printing.
It is evident from the above discussion that there are various problems associated with inkjet printing organic materials for manufacturing OLEDs.
It is an aim of embodiments of the present invention to solve, or at least reduce, one or more of the problems outlined above by adapting compositions for ink jet printing organic light-emissive materials for light-emissive devices.
While the aforementioned discussion has been directed at inkjet printing it should be noted that at least some of these problems may occur when using any deposition technique which involves passing a composition through one or more openings under pressure to deposit the composition. Examples of such deposition techniques include inkjet printing, nozzle printing and spray coating.